Highway guardrail health monitoring and early warning system based on big data

Abstract

The invention relates to the technical field of health monitoring, in particular to a highway guardrail health monitoring and early warning system based on big data, which comprises a sensing data acquisition and monitoring module, a spatio-temporal data fusion and health state judgment module, a damage development trend monitoring module, a repair priority ranking module and a risk early warning and response module. According to the invention, through automatic data acquisition and analysis, the health condition of the highway guardrail is monitored, the sampling frequency is dynamically adjusted according to environment change, abnormal data is removed, noise is reduced, health state judgment is combined with regional data weight analysis, the health condition of the guardrail is evaluated, and a health judgment result is formed; according to damage development trend monitoring, the influence of a damage aggravation area and external factors is identified, a risk area is preferentially processed, a repair task is reasonably arranged, a risk early warning mechanism is combined with real-time data and repair priority evaluation, emergency response is started, and the response capacity and decision-making efficiency of the system are improved.

Description

Technical Field

[0001] This invention relates to the field of health monitoring technology, and in particular to a highway guardrail health monitoring and early warning system based on big data. Background Technology

[0002] The field of health monitoring technology involves the real-time monitoring and assessment of equipment, the environment, and human health status through various means and technologies. This includes sensor technology, data acquisition and processing, big data analysis, and intelligent early warning. Core aspects of this field include accurate data acquisition, real-time monitoring, effective risk warning, and decision support. Key methods include installing multiple types of sensors to collect data, transmitting information via wireless communication technology, and using data analysis algorithms to process and analyze the monitoring data in order to promptly identify potential problems and risks, providing scientific evidence and decision support. In the field of health monitoring, with technological advancements, the intelligence, real-time performance, and automation levels of monitoring systems are continuously improving, leading to their widespread application in environmental monitoring, public safety, and healthcare, among other fields. Traditional highway guardrail health monitoring and early warning systems based on big data refer to systems that use big data technology to monitor and warn of the status of highway guardrails in real time. These systems collect structural information, environmental data, and traffic conditions from installed sensors, and then use big data analytics to process the collected data, assessing the health status of the guardrails in real time and identifying potential faults and safety hazards. Traditional highway guardrail monitoring typically relies on manual inspections and periodic checks, which suffer from problems such as untimely response, limited inspection coverage, and high costs. To address these issues, this patent utilizes automated data collection and analysis, employing big data technology for real-time monitoring and early warning of guardrail status, thereby improving guardrail safety and management efficiency.

[0003] Current technologies rely on manual inspections and periodic checks, resulting in slow response times and limited coverage. Manual inspections cannot monitor the condition of guardrails in real time, cannot quickly identify potential hazards, and the frequency of inspections cannot meet dynamically changing needs, leading to delays in problem detection. Periodic checks depend on manual operation, which carries the risk of omissions or errors and cannot efficiently handle the high-frequency or extreme environment health monitoring needs, resulting in some potential hazards not being identified in a timely manner. The inspection process lacks automation and data support, and cannot provide real-time feedback, affecting the timeliness and accuracy of decision-making, increasing repair costs and traffic safety risks. Summary of the Invention

[0004] To address the technical problems existing in the prior art, this invention provides a highway guardrail health monitoring and early warning system based on big data. The technical solution is as follows: On the one hand, a highway guardrail health monitoring and early warning system based on big data is provided, which includes: The sensor data acquisition and monitoring module acquires temperature, pressure, displacement, and vibration data, sets the sampling frequency, removes abnormal data, and suppresses interference through low-pass filtering to obtain health monitoring data. The spatiotemporal data fusion and health status determination module, based on the health monitoring data, analyzes regional characteristics through spatial distribution and weight information to determine the health status of the guardrail and obtain the health status of the guardrail. Based on the health status of the guardrail, the damage development trend monitoring module analyzes the health changes in the monitoring area, identifies trend differences, delineates areas of increased damage, assesses the impact of external factors on the damage, determines the damage risk, and obtains the damage development trend results. Based on the damage development trend results, the repair priority ranking module assesses the environmental impact from a traffic safety perspective, analyzes the urgency of repair tasks according to factors such as the degree of damage, environmental impact weight, and repair resources, and derives the repair priority results. The risk warning and response module assesses the urgency, risk level, resources, and environment of the task based on the repair priority, determines whether to initiate an emergency response and issue a warning, reassesses the risk, and obtains a risk warning result.

[0005] As a further aspect of the present invention, the health monitoring data includes temperature data, pressure data, displacement data, and vibration data; the data completeness and continuity include missing data segments and abnormal segments; the noise reduction effect after data filtering specifically refers to noise elimination and interference reduction using Kalman filtering; the guardrail health determination status includes the health status and the health status of the monitored area; the health determination analysis includes the impact of regional data on monitoring results, data weight allocation, data mapping, and alignment of time and spatial information; the damage development trend results include areas of increased damage, health change trajectories obtained through time series analysis, damage risk calculated based on a multivariate regression model, and the specific impact mechanism of external factors on damage obtained through environmental parameter analysis; the repair priority results include the repair task sequence, the urgency of repair tasks, repair demand analysis, and repair task priority; and the risk warning results include warning notification, emergency response activation, risk assessment, and reassessment of feedback information.

[0006] As a further aspect of the present invention, the mapping area refers to the area to which the monitoring data belongs, identified according to geographical features, and the monitoring data is mapped to a specific geographical range according to the location relationship in the spatiotemporal data fusion and health status determination module. The data weight refers to the proportion assigned to differential monitoring data during the health status determination process, based on sampling frequency, signal integrity, and regional criticality.

[0007] As a further aspect of the present invention, the trend difference refers to the damage development trend monitoring module, which identifies the direction and magnitude differences in health changes by comparing health data results over different time periods, thereby determining the development trend of guardrail damage. The emergency response refers to the process in the risk warning and response module where, when the risk assessment results indicate the existence of an urgent hidden danger, the emergency response procedure is initiated according to the set standards, an early warning notice is issued, and on-site response and emergency handling measures are organized and implemented.

[0008] As a further aspect of the present invention, the sensor data acquisition and monitoring module includes: The data acquisition submodule acquires temperature, pressure, displacement, and vibration data of the guardrail, sets different sampling frequencies according to changes in environmental parameters, performs data acquisition, and obtains raw sensor data by synchronizing data in time and space sequence. The data cleaning submodule, based on the original sensor data, determines the integrity and continuity of the data, removes missing data and abnormal segments, and reduces noise and interference through filtering to obtain a cleaned data stream. The filtering submodule analyzes the health status information of the guardrail based on the cleaned data stream, monitors dynamic changes and identifies key features, and analyzes parameters such as temperature, pressure, displacement, and vibration to obtain health monitoring data.

[0009] As a further aspect of the present invention, the spatiotemporal data fusion and health status determination module includes: The time synchronization submodule synchronizes each data item based on the health monitoring data using timestamps, calibrates the time information of different data sources, and compares and analyzes data items within a unified time range to obtain time-aligned data. The regional data mapping submodule identifies the monitoring area corresponding to each data item based on the time-aligned data, maps the data to the corresponding area through regional features, and points the data to the corresponding area according to the geographical location to obtain regionalized monitoring data. The health status determination submodule assigns weights to each data item based on the regionalized monitoring data, analyzes the impact of regional weights on health monitoring results, and determines the health status of each monitoring area by comparing data from multiple regions, thus obtaining the health status of the guardrail.

[0010] As a further aspect of the present invention, the process of the time synchronization submodule performing timestamp calibration based on health monitoring data is specifically as follows: comparing multiple data with the same time reference, identifying time differences between different data sources based on the comparison results, and correcting the time order of the data based on the time differences; The process of the regional data mapping submodule for regional identification based on time-aligned data is as follows: the monitoring data is corresponding to the geographical feature information. When the location information of the monitoring data exceeds the boundary of the regional feature, the data is updated according to the spatial adjacency relationship of the regional range. The process of the health status determination submodule in analyzing regional monitoring data is as follows: data weights are set according to sampling frequency and signal integrity, and the role of data from different regions in the overall assessment is allocated through weight ratio.

[0011] As a further aspect of the present invention, the damage development trend monitoring module includes: The health change trend identification submodule extracts health data from different time periods based on the health determination status of the guardrail, analyzes the health changes in the monitored area, compares the determination results of different time periods, identifies the change trend through data differences, and obtains health change trend information. The damage aggravation area delineation submodule identifies and delineates areas of aggravated damage based on the health change trend information, analyzes the health changes in multiple areas, and obtains damage aggravation area data. The damage risk assessment submodule analyzes the impact of external factors on damage changes based on the data of the damage aggravation area, assesses the health change trajectory, and determines the damage risk based on the analysis results, thereby obtaining the damage development trend results.

[0012] As a further aspect of the present invention, the repair priority sorting module includes: The damage urgency assessment submodule assesses the damage level and development rate of the area based on the damage development trend results, determines the urgency of the repair task by comparing the degree of damage aggravation, and obtains environmental data. The environmental impact analysis submodule analyzes the impact of external factors on damage changes based on the environmental data, and judges the impact of environmental factors on remediation needs through damage development trend data, thereby obtaining the environmental factor impact results. The repair task sorting submodule sets the task order based on the environmental factors, analyzes the urgency, influencing factors and repair priority of the tasks, sets the priority of the repair tasks, and obtains the repair priority result.

[0013] As a further aspect of the present invention, the risk warning and response module includes: Based on the repair priority results, the risk assessment submodule analyzes the associated data of each repair task, assesses the risk of the task, analyzes the urgency and available response time of the task, determines the risk level of task execution, and obtains the risk assessment result. The emergency response determination submodule determines whether to initiate the emergency response process based on the risk assessment results and the urgency of the task. According to the set response criteria, it analyzes the repair tasks that need to be handled immediately and obtains the determination result of whether to initiate the emergency response. The early warning feedback processing submodule reassesses the early warning based on the determination result of whether to initiate an emergency response and the feedback information, changes the risk assessment result through real-time data, and fixes the task priority according to the feedback information to obtain the risk early warning result.

[0014] Compared with the prior art, the advantages and positive effects of the present invention are as follows: In this invention, the health status of highway guardrails is monitored through automated data collection and analysis. The system dynamically adjusts the sampling frequency according to environmental changes, removes abnormal data and reduces noise. The health status determination combines regional data weight analysis to assess the health status of the guardrails and form a health determination result. The damage development trend monitoring identifies areas of increased damage and the influence of external factors, prioritizes the treatment of risk areas, and rationally arranges repair tasks. The risk warning mechanism combines real-time data and repair priority assessment to initiate emergency response, thereby improving the system's responsiveness and decision-making efficiency. Attached Figure Description

[0015] To more clearly illustrate the technical solutions in the embodiments of the present invention, the accompanying drawings used in the description of the embodiments will be briefly introduced below. Obviously, the accompanying drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0016] Figure 1 This is a system flowchart of the present invention; Figure 2 This is a system block diagram of the present invention; Figure 3 This is a flowchart of the sensor data acquisition and monitoring module in this invention; Figure 4 This is a flowchart of the spatiotemporal data fusion and health status determination module in this invention; Figure 5 This is a flowchart of the damage development trend monitoring module in this invention; Figure 6 This is a flowchart of the priority sorting module in this invention; Figure 7 This is a flowchart of the risk warning and response module in this invention. Detailed Implementation

[0017] The technical solution of the present invention will now be described with reference to the accompanying drawings.

[0018] In embodiments of the present invention, words such as "exemplarily," "for example," etc., are used to indicate that something is an example, illustration, or description. Any embodiment or design described as "exemplary" in the present invention should not be construed as being more preferred or advantageous than other embodiments or designs. Specifically, the use of the word "exemplary" is intended to present the concept in a concrete manner. Furthermore, in embodiments of the present invention, the meaning expressed by "and / or" can be both, or either one.

[0019] In the embodiments of this invention, the terms "image" and "picture" may sometimes be used interchangeably. It should be noted that, without emphasizing the distinction between them, they convey the same meaning. Similarly, the terms "of," "corresponding (relevant)," and "corresponding" may sometimes be used interchangeably. It should be noted that, without emphasizing the distinction between them, they convey the same meaning.

[0020] In this embodiment of the invention, sometimes a subscript such as W1 may be written in a non-subscript form such as W1. When the difference is not emphasized, the meaning they express is the same.

[0021] To make the technical problems, technical solutions and advantages of the present invention clearer, a detailed description will be given below in conjunction with the accompanying drawings and specific embodiments.

[0022] This invention provides a highway guardrail health monitoring and early warning system based on big data, such as... Figure 1-2 The diagram shown illustrates a highway guardrail health monitoring and early warning system based on big data. The system includes: The sensor data acquisition and monitoring module acquires temperature, pressure, displacement and vibration data of the highway guardrail. It sets the sampling frequency according to the changes in environmental parameters, collects guardrail health data, analyzes the completeness and continuity of the data, removes missing and abnormal segments, and filters the data to reduce noise and weaken interference, thus obtaining health monitoring data. The spatiotemporal data fusion and health status determination module is based on health monitoring data, aligns time and spatial information, synchronizes each data item according to timestamp, compares regional data within a unified time range, points data to corresponding regions and maps data according to geographical location, assigns data weights according to regional characteristics, analyzes the impact of weighted regional data on monitoring results, determines the health status of each monitoring region, and obtains the guardrail health status. The damage development trend monitoring module analyzes the health changes in the monitored area based on the health status of the guardrail, compares the judgment results of different time periods to identify the trend of change, delineates the area of ​​damage aggravation, analyzes the impact of external factors on the damage change process based on environmental data, assesses the health change trajectory, determines the damage risk, and obtains the damage development trend results. The repair priority ranking module analyzes the urgency of multiple repair tasks based on the damage development trend results, assesses the damage level and development rate of the area, analyzes the impact of environmental factors on repair needs from the perspective of traffic safety, determines whether to enter the immediate repair process, sets the task order according to the urgency of the repair tasks, and obtains the repair priority results. The risk warning and response module analyzes the risks of multiple repair tasks based on the repair priority results, determines whether to initiate an emergency response based on the assessment of urgency and available response time, issues warning notifications in actual scenarios, and re-evaluates the warnings based on feedback information to obtain risk warning results.

[0023] Health monitoring data includes temperature, pressure, displacement, and vibration data. Data completeness and continuity include missing and abnormal data segments. The noise reduction effect after data filtering specifically refers to noise elimination and interference reduction using Kalman filtering. The health status of the guardrail includes its health status and the health status of the monitored area. Health assessment analysis includes the impact of regional data on monitoring results, data weight allocation, data mapping, and alignment of time and space information. Damage development trend results include areas of increased damage, health change trajectories obtained through time series analysis, damage risk calculated based on a multivariate regression model, and the specific impact mechanism of external factors on damage obtained through environmental parameter analysis. Repair priority results include the order of repair tasks, urgency of repair tasks, repair needs analysis, and repair task priority. Risk warning results include warning notification, emergency response activation, risk assessment, and reassessment of feedback information.

[0024] Specifically, such as Figure 2 , 3 As shown, the sensor data acquisition and monitoring module includes: The data acquisition submodule acquires temperature, pressure, displacement, and vibration data of the guardrail, sets different sampling frequencies according to changes in environmental parameters, performs data acquisition, and obtains raw sensor data by synchronizing data in time and space sequence. First, environmental parameters must be collected, and the data collection frequency adjusted accordingly. For example, temperature changes are predicted to affect the pressure or displacement of the guardrail. Therefore, the sampling frequency is adjusted based on parameters such as temperature and humidity according to a set algorithm. If the temperature changes drastically, the sampling frequency is increased to capture the impact of temperature changes on pressure, displacement, and other data more precisely. Conversely, the frequency is decreased. For example, if the temperature changes by more than 5 degrees Celsius per hour, the sampling frequency is increased from every 10 seconds to every 5 seconds; if the temperature change is small, the sampling frequency returns to the normal frequency. In addition, data collection not only depends on the temporal sequence but also requires spatial synchronization. To ensure the consistency of data collected by multiple sensors, GPS signals or other positioning methods are used to ensure the synchronization of sensor positions. Taking pressure data from a single target spatial location as an example, the displacement and pressure sensors of a single sensor are set to be 10 meters apart. The synchronization mechanism ensures that these two types of data are correctly correlated at the same time point, thereby ensuring that the final raw data has spatial and temporal consistency, ultimately yielding the raw sensor data.

[0025] The data cleaning submodule, based on the original sensor data, determines the integrity and continuity of the data, removes missing data and abnormal segments, and reduces noise and interference through filtering to obtain a cleaned data stream. First, the raw data needs to be inspected to determine if there are missing or outlier values. Specifically, missing data is usually identified by comparing the timestamp with the actual data collection time. If data was not collected in a timely manner within a single time period or if there are significant jumps between data points, the data point can be marked as missing data and processed iteratively. Furthermore, if the number of missing data points exceeds a set threshold, the data segment needs to be removed from the data stream. Additionally, it is necessary to determine if there are anomalous segments. Anomalous segments typically exhibit significant inconsistencies with surrounding data, including a single data point's temperature value being much higher than the normal range, presumably due to sensor malfunction or external interference. In this case, the data point is first compared with several points before and after it to check for similar fluctuation trends or abrupt changes. If none are found, the data is determined to be anomalous and removed. Next... To ensure the quality of the final data stream, noise and interference filtering is performed. Noise typically manifests as high-frequency fluctuations or instantaneous abnormal fluctuations in the data, including sudden data changes when sensors are subjected to electromagnetic interference. To reduce these effects, low-pass filters can be used. This involves setting a maximum frequency threshold to remove fluctuations exceeding this frequency, thus smoothing the data and reducing high-frequency noise. The filter can be implemented by weighted averaging of the neighboring data for each data point, or by converting the data to the frequency domain using a fast Fourier transform for filtering, and then converting it back to the time domain to obtain the processed data. In addition, for data fluctuations over long periods, if the fluctuation amplitude is small and within the normal range of environmental changes, a minimum fluctuation threshold can be set during filtering to avoid mistakenly removing normal environmental changes. Through these operations, a cleaned data stream is finally obtained.

[0026] The filtering submodule analyzes the health status information of the guardrail based on the cleaned data stream, monitors dynamic changes and identifies key features, and analyzes parameters such as temperature, pressure, displacement, and vibration to obtain health monitoring data. First, the data after cleaning needs to be analyzed item by item. Parameters such as temperature, pressure, displacement, and vibration are monitored as differentiated dimensions of health status. For example, temperature fluctuations reflect the thermal expansion and contraction of the guardrail under changes in ambient temperature; pressure changes reflect the deformation of the guardrail under external forces; displacement data provides the overall displacement status of the guardrail; and vibration data reflects the response of the guardrail to external vibration sources. By analyzing these data item by item, the dynamic changes of the guardrail can be monitored. For example, if the temperature change is greater than 5°C within a certain period, it indicates a rapid change in the environment, requiring iterative monitoring of the temperature and pressure relationship of the guardrail to determine whether pressure instability occurs due to excessive temperature differences. If the vibration amplitude exceeds the preset range, including a vibration amplitude greater than 0.5mm, it is estimated that the guardrail has been subjected to abnormal external force. It is necessary to iteratively check whether there is structural damage or external interference. In order to identify key features, data can be filtered and analyzed by setting differentiated thresholds. For example, a temperature change threshold of ±3°C can be set. If it exceeds this threshold, it is considered a key change and the data point will be marked. The relationship between the data and parameters such as pressure, displacement, and vibration will be iteratively calculated and compared to determine whether there are potential safety hazards. In this process, historical data is compared, the changing trends of multiple parameters are analyzed, the correlation between multiple parameters and key change points are identified, thereby determining the health status of the guardrail and finally obtaining health monitoring data.

[0027] Specifically, such as Figure 2 , 4 As shown, the spatiotemporal data fusion and health status determination module includes: The time synchronization submodule synchronizes each data item based on health monitoring data and timestamps it, calibrates the time information of different data sources, and compares and analyzes data items within a unified time range to obtain time-aligned data. First, the timestamps of each data item need to be extracted and compared to ensure that the time information from all data sources is processed under a unified standard. This includes situations where there might be a time discrepancy between the estimated temperature data from sensor A and the estimated pressure data from sensor B, caused by differences in device startup times or network latency. In such cases, the timestamps of these two data sources are compared. If a discrepancy is found, it needs to be corrected to align the timestamps of the two data sources. The specific execution process is as follows: First, extract the timestamps of the data from sensors A and B, calculate the time difference between them, and set the timestamp of the temperature data from sensor A as T1 and the timestamp of the pressure data from sensor B as T2. If the time difference between T1 and T2 is 5 seconds, interpolation or delay processing is needed to adjust the timestamp of one of the data sources. To ensure consistency with another data source, a common practice is to use linear interpolation to calculate the required timestamp values ​​based on adjacent sampling points, thereby synchronizing the timestamps. After data calibration, all data items are aligned according to a unified timestamp and merged into a single data stream. This includes sampling frequencies of 10 seconds for temperature data, 15 seconds for pressure data, and 5 seconds for vibration data. Data with differing frequencies need to be adjusted to the same time scale based on the timestamps. To achieve this, all data undergoes interpolation to align data points to the same time interval. This ensures that accurate values ​​of multiple data types are obtained simultaneously, avoiding comparison and analysis errors caused by differences in data frequency. In this way, the time-aligned data is finally obtained.

[0028] The regional data mapping submodule identifies the monitoring area corresponding to each data item based on the time-aligned data, maps the data to the corresponding area through regional features, and points the data to the corresponding area according to the geographical location to obtain regionalized monitoring data. First, each set of time-aligned data needs to be analyzed to identify the correlation between the data and the target monitoring area. This step typically relies on the geographic location information of the monitoring equipment, including GPS coordinates. These coordinates determine the geographical location of the data collection point. When a single sensor is used to monitor a guardrail, it provides data such as temperature, pressure, and displacement. Since the sensor is located in a single, specific location, the latitude and longitude of that location are first obtained using the sensor's GPS information. Then, by comparing this location with a pre-defined monitoring area division, it is determined whether the location belongs to the monitoring area of ​​a single target. For example, if the coordinates of the single sensor are [value missing], by finding the monitoring area boundary information, it can be identified as belonging to area A. At this point, the data is mapped to area A, ensuring its [value missing]. Associated with this region, the same mapping operation is then performed on data from other monitoring areas based on the geographical location of each data point. For scenarios where data is relatively dispersed, if sensors are distributed across multiple regions, the region to which each data point belongs can be determined by setting geographical partitioning criteria, including using latitude and longitude ranges or defining regional boundaries using landmarks within the region. For example, if the latitude and longitude range of region A is specified, and the sensor is located within this range, the data will be assigned to region A. Similarly, for data points within region B, it will be determined whether they are located within the range of region B. If the condition is met, the data will be assigned to region B. Through these steps, the data will be classified and labeled according to its corresponding monitoring region, ultimately resulting in regionalized monitoring data.

[0029] The health status determination submodule assigns weights to each data item based on regional monitoring data, analyzes the impact of regional weights on health monitoring results, and determines the health status of each monitoring area by comparing data from multiple regions, thus obtaining the health status of the guardrail. First, a weight needs to be assigned to each data point based on the criticality of each monitoring area and the quality of the monitored data. For example, monitoring data from a single area, A, indicates drastic temperature and pressure changes, suggesting a significant risk to the guardrails in that area. In contrast, monitoring data from area B is relatively stable. Therefore, the weight of data from area A is estimated to be higher, at 0.7, while the weight of data from area B is set at 0.3. This weight allocation can be determined based on factors such as the risk assessment of the area, the reliability of historical data, and the complexity of the monitoring environment. Next, a comprehensive analysis of the monitoring data from each area is conducted, comparing the trends in regional data and their correlation with health status. For instance, the temperature variation range in area A is ±2°C, while in area B it is ±1°C. Considering the weight settings for each area, the trend prediction for area A has a greater impact on the overall health assessment. The impact assessment involves weighting the health data from all areas. If data from area A indicates potential damage while area B shows no abnormalities, the overall health status will still be affected by area A. The final health status assessment will indicate a risk in that area. Then, a comparative analysis will be performed based on the weighted regional data to determine the health status of each monitored area. For example, if the health status of area A and area B is "damaged" and "normal" respectively, and area A has a higher weight, the overall health status of the guardrail will be assessed as "damaged." This process ensures a more accurate health assessment by comprehensively considering the weight of each area's data and the health status it reflects. Simultaneously, the weights and health status assessments will be updated periodically based on real-time data changes to ensure the timeliness and accuracy of the monitoring results. Through this series of operations, the final health status of the guardrail is determined.

[0030] Specifically, such as Figure 2 , 5 As shown, the damage development trend monitoring module includes: The health change trend identification submodule extracts health data from different time periods based on the health status of the guardrail, analyzes the health changes in the monitored area, compares the judgment results of different time periods, identifies the change trend through data differences, and obtains health change trend information. First, based on the health status assessment of each monitored area, health data for differentiated time periods needs to be extracted. This includes recording and extracting health status data for differentiated monitored areas such as Area A, Area B, and Area C within two differentiated time periods, T1 and T2. This data includes changes in parameters such as temperature, pressure, and displacement. By matching these data with timestamps, corresponding health data points within these time periods are extracted. For example, if the health status of Area A is "normal" at time T1 but "damaged" at time T2, the change in the health status of that area will be marked, and the associated data points will be used as health data for the differentiated time periods. Next, the health data for the differentiated time periods will be analyzed, including the health changes of the monitored areas. This analysis not only compares changes in health status but also includes detailed recording of trend changes in associated parameters. For example, if the temperature in Area A is 20°C at T1 and 25°C at T2, and the pressure changes from 50Pa to 70Pa, this data needs to be analyzed. To analyze whether the increase in temperature and pressure is related to the change in health status in region A, this data will be combined with changes in health status. Similar analyses will be performed on data from other regions to identify similar trends or changes, and to determine if other regions are also experiencing changes in health status, or if changes in the health status of a single monitored region have affected other regions. Then, the results of these time-period differences will be compared. For example, at time T1, region A's health status is "normal," and region B's is "damaged," while at time T2, region A changes to "damaged," and region B changes to "recovered to normal." The health changes in these two time periods will be compared, and trends will be identified based on data differences. For instance, if the temperature increase in region A exceeds a set threshold, while the temperature fluctuation in region B is smaller within the same time period, it can be determined that the health change in region A is related to the temperature increase. Through these comparisons and by comprehensively referring to the health status change trends of all monitored regions, the final health change trend information is obtained.

[0031] The damage aggravation area delineation submodule identifies and delineates areas of aggravated damage based on health change trend information, analyzes the health changes in multiple areas, and obtains damage aggravation area data. First, it's necessary to extract health data change records from multiple monitoring areas based on health change trend information. These records include changes in various parameters such as temperature, pressure, and displacement. Time series analysis will be performed on this data to identify drastic changes in the health status of each area within a target time period. For example, if the temperature in area A rises from 30°C to 40°C in the past 24 hours, accompanied by an increase in pressure from 50Pa to 80Pa, while the temperature in area B remains relatively stable at 32°C to 33°C, the change in area A will be considered an abnormal fluctuation. Iterative analysis will then be conducted to determine if there is a trend of worsening damage in this area. Next, the health changes in multiple areas need to be compared to determine which areas show the most significant worsening damage. Specifically, the rate of change and the amplitude of fluctuations in each area will be compared. This includes cases where the health data of areas A and B both fluctuated within the same time period, but... If the variation in region A is significantly greater than that in region B, and the fluctuation trend persists for a longer period, then the damage aggravation in region A will be considered more severe. To more accurately identify areas of aggravated damage, threshold judgment criteria are set, including: if the temperature change exceeds 5°C, the pressure change exceeds 30Pa, and the duration exceeds 3 hours, it will be considered that the region has shown signs of aggravated damage. Subsequently, data will be filtered according to these criteria to identify all areas of aggravated damage that meet the conditions. For example, if the temperature and pressure in region A both exceed the set thresholds during the monitoring period, and the fluctuation amplitude exceeds the predetermined standard, region A will be marked as an area of ​​aggravated damage, and its data will be included in iterative analysis. Through these operations, areas with a trend of aggravated damage within the target time period can be accurately identified, providing strong data support for subsequent maintenance and repair, and ultimately obtaining data on areas of aggravated damage.

[0032] The injury risk assessment submodule analyzes the impact of external factors on injury changes based on data from areas of increased injury, assesses the trajectory of health changes, determines the injury risk based on the analysis results, and obtains the injury development trend results. First, a detailed analysis of the data from the areas of increased damage is needed to identify the specific impact of external factors on damage changes. This includes analyzing the relationship between temperature and pressure changes in area A, where temperature and pressure gradually increase, while the environmental temperature in area B changes relatively little. By comparing data and integrating external environmental information, we can derive some predicted external factors that exacerbate the damage process in area A. These external factors include the predicted increase in temperature or traffic leading to increased stress on the guardrails, thus accelerating the damage progression. Next, the health change trajectory will be assessed, analyzing the trends in health status changes across multiple areas within a target time period. This includes the gradual deterioration of the health status in area A over the past week, with temperature rising from 30°C to 40°C and pressure from 50 Pa. An increase to 80 Pa indicates that the damage is worsening. The rate of change in health in region A is calculated, and these changes are analyzed to determine if they align with the expected trajectory of worsening damage. The trends in other regions are compared to check for similar patterns. If the rate of change in region A is significantly higher than in other regions, iterative verification can be performed to confirm that the damage in that region is accelerating. Then, the damage risk is determined based on the analysis results. Specifically, if the damage changes in region A meet set threshold conditions, such as temperature changes exceeding 5°C or pressure changes exceeding 30 Pa, and the changes last for a prolonged period, the damage risk in that region is assessed as high. The assessment of damage risk relies not only on the current health change trajectory but also on historical data and external environmental factors to obtain a more accurate assessment, ultimately leading to the damage development trend.

[0033] Specifically, such as Figure 2 , 6 As shown, the repair priority sorting module includes: The damage urgency assessment submodule assesses the damage level and rate of development of a region based on the damage development trend results. By comparing the degree of damage aggravation, it determines the urgency of the repair task and obtains environmental data. First, based on the damage development trend results, the damage level of multiple monitoring areas needs to be assessed. This includes setting area A's temperature to rise from 30°C to 45°C and its pressure to increase from 50 Pa to 100 Pa over a past period, while area B's temperature change is relatively small, only ±2°C, and its pressure remains stable at 50 Pa. Based on these changes, area A is judged to have a significantly higher damage level than area B. The assessment criteria for area A's damage level include temperature changes exceeding 5°C and pressure fluctuations exceeding 30 Pa, which can be considered areas with a higher damage level. For area B, due to the smaller changes, the assessment result is a lower damage level. Next, the rate of damage aggravation needs to be calculated to determine the speed of damage progression. This includes calculating the rate of temperature change in area A over a time period and the rate of pressure change over a time period. By setting the time period to 24 hours, the rates of change in temperature and pressure can be calculated and compared with those in other areas. For example, if the rate of temperature change in area A is much higher than that in area B, the damage in area A is considered to be progressing faster, thus giving a higher assessment of the urgency of the repair task. Then, the urgency of the repair task is determined by comparing the degree of damage aggravation. If the rate of damage progression in an area is significantly higher than that in other areas, and the impact of the damage is more severe, the repair task in that area will be marked as "urgent," and the repair priority will be noted in the environmental data. For example, if the damage changes in area A have reached a risk level, it will be determined that the area needs to be repaired as soon as possible based on these factors, while other areas, including area B, can be judged as having "normal" or "lower urgency" repair tasks. Finally, the environmental data is obtained.

[0034] The environmental impact analysis submodule analyzes the impact of external factors on damage changes based on environmental data, and judges the impact of environmental factors on remediation needs through damage development trend data, thus obtaining the results of environmental factor impact. First, it is necessary to collect and analyze environmental data related to the damage, including temperature, humidity, wind speed, and precipitation. These external factors are expected to directly affect the changes in guardrail damage. For example, if the temperature in area A fluctuates significantly over a period of time, rising from 25°C to 40°C, accompanied by strong winds and heavy rain, while the temperature in area B remains relatively stable at only 30°C, the impact of these external factors on the aggravation of damage will be analyzed. For area A, due to the rapid temperature change and extreme weather conditions, it is estimated that the aging of the guardrail material or structural damage is accelerated, leading to aggravated damage. These external factors need to be correlated with the health data of area A to calculate the contribution of external factors to the damage change. Next, the impact of these external factors will be iteratively analyzed based on the damage development trend data. This includes assessing the degree of influence of temperature changes on damage development, given that the damage in area A significantly worsened during the temperature fluctuation period. The temperature in area A will be set to reflect the past 48 hours... The temperature fluctuated by 10°C, while the temperature change in area B was smaller, but the degree of damage did not change much. This leads to the conclusion that temperature changes have a significant impact on the damage in area A. Furthermore, other external factors, including wind speed and precipitation, are combined to iteratively analyze their estimated impact on damage aggravation. For example, heavy rain is predicted to cause damage to the guardrail structure, and high wind speeds are predicted to increase external forces, thus accelerating the damage process. Subsequently, these impacts are quantified to determine the influence of environmental factors on repair needs. For instance, if the degree of damage in area A increases by 30% under the influence of the target external factors, while the degree of damage in area B only increases by 5%, the urgency of repair needs can be determined by calculating the impact ratios of these changes. If the aggravation of damage in area A is closely related to external environmental factors, repairs need to be carried out as soon as possible to avoid iterative deterioration. Based on this data, environmental factor impact results are generated, and it is determined which external factors require special attention, ultimately yielding the environmental factor impact results.

[0035] The task ordering submodule sets the task order based on the impact of environmental factors, analyzes the urgency, influencing factors and repair priority of the tasks, sets the priority of the repair tasks, and obtains the repair priority results. First, key data for multiple remediation tasks needs to be extracted from the environmental factors' impact results. This data includes the specific degree of impact of environmental factors on multiple monitoring areas. For example, damage in area A is significantly affected by extreme weather (including high temperatures and heavy rain), while environmental conditions in area B are relatively stable. Tasks will be prioritized based on these influencing factors. For instance, in area A, the environmental factors exacerbate the damage, making the remediation tasks more urgent, while in area B, due to smaller temperature changes and limited precipitation, the tasks are less urgent. At this point, an initial urgency assessment will be assigned to each remediation task based on the impact of external environmental factors on the damage. Next, the urgency of each task will be analyzed, including a comprehensive assessment of environmental factors, damage progression rate, and the degree of damage exacerbation. The remediation tasks in area A are prioritized due to external weather factors, while the remediation tasks in area B are estimated to be affected by weather conditions. The impact of the weather is relatively small, so its urgency is low. To accurately assess the urgency, multiple criteria were set, including areas with temperature changes exceeding 5°C and precipitation exceeding 100mm, where the urgency of the repair task is high and its repair priority is set to 1, indicating that it needs to be dealt with immediately; if the precipitation and temperature changes are small, the repair priority is 3, indicating that repair can be arranged later. Then, the priority of the repair tasks is set according to the urgency and influencing factors. By comparing the repair tasks in multiple areas, the priority ranking of differentiated tasks can be calculated. For example, the repair task priority of area A is set to 1, the repair task priority of area B is set to 3, and if the repair task of area C is less affected by external climate and the damage is less, its repair priority is estimated to be set to 2. Based on these assessment results, a final priority list of repair tasks is generated, and the final repair priority result is obtained.

[0036] Specifically, such as Figure 2 , 7 As shown, the risk warning and response module includes: The risk assessment submodule analyzes the associated data of each remediation task based on the remediation priority results, assesses the risk of the task, analyzes the urgency and available response time of the task, determines the risk level of task execution, and obtains the risk assessment results. First, it's necessary to extract the associated data for each repair task from the repair priority results. This data typically includes the urgency of the repair task, the damage level of the area where the task is located, the impact of environmental factors, and the available resources and response time. For example, the repair task for area A is rated as priority 1, indicating high urgency, and this area is also affected by extreme weather. The repair task for area B is rated as priority 3, indicating lower urgency. We will first analyze the associated data for the repair task for area A, focusing on the degree of damage aggravation, weather changes, and the adequacy of repair resources in the area. For area B, due to less damage and a stable external environment, it will be assessed as a low-risk task. Next, we analyze the urgency and available response time of the tasks. The repair task for area A has a high urgency, but due to the impact of extreme weather, the response time is short, and it is estimated to need to be completed within 24 hours. While the repair task for area B has a low priority, the estimated repair time is more lenient. The process includes a 72-hour timeframe, iteratively assessing the risks of repair tasks based on their urgency and available response time. For example, if a repair task in Area A needs to be expedited due to weather changes, the predictability of task completion and the associated risks will be determined by analyzing weather forecasts and the availability of repair resources. For Area B, although the urgency is lower, the risk level will be assessed as medium due to the longer response time. Subsequently, the execution risk of each repair task is assessed, focusing on the following aspects: whether the resources required for the repair task are sufficient, whether the repair equipment is available, whether personnel are in place, and whether environmental conditions are expected to affect the repair work. For instance, if a repair task in Area A requires special materials or equipment that are unavailable at the moment, the risk level of that task will be assessed as "high," and resource allocation for Area A will be prioritized. The risk level of Area B will also be assessed as "low" or "medium" based on its resource status and environmental conditions, ultimately yielding the final risk assessment result.

[0037] The emergency response determination submodule determines whether to initiate the emergency response process based on the risk assessment results and the urgency of the task. According to the set response standards, it analyzes the repair tasks that need to be handled immediately and obtains the determination result of whether to initiate the emergency response. First, based on the risk assessment results, the urgency of each remediation task needs to be evaluated. Remediation tasks in Area A are deemed highly urgent due to extreme weather, while tasks in Area B, although requiring remediation, have lower urgency due to less damage and a more stable environment. Based on these assessments, the urgency of each task is categorized: high-urgency tasks are designated "urgent," and low-urgency tasks are designated "normal." Next, combining the task's urgency with environmental data, it is determined whether to activate the emergency response procedure. In Area A, the risk level of the remediation task is assessed as high, and its urgency level is assessed as "urgent." Therefore, the emergency response procedure will be activated according to the established response criteria. The criteria for initiating an emergency response include: the urgency of the repair task exceeds a certain threshold, or the risk level of the task exceeds a single benchmark value. By comparing the urgency of the tasks with the risk assessment results, it is possible to accurately determine which tasks require immediate response. Then, the repair tasks that need immediate attention are analyzed to identify their priorities. Combined with external factors such as repair resources, personnel arrangements, and weather conditions, it is ensured that the emergency response process can be quickly initiated and effectively executed. For example, if the repair task in Area A requires special equipment or materials, and these resources are available at the current time and the task is urgent, this task will be assigned the highest priority, and the emergency response process will be initiated to ensure that the repair work can begin as soon as possible. Finally, a decision is made on whether to initiate an emergency response.

[0038] The early warning feedback processing submodule reassesses the early warning based on the determination result of whether to activate the emergency response and the feedback information. It changes the risk assessment result through real-time data and fixes the task priority according to the feedback information to obtain the risk warning result. First, based on the determination of whether the emergency response has been activated, and combined with feedback information, the warning level is reassessed. In Area A, the emergency response process was activated due to weather conditions, and feedback data shows a continued rise in temperature and drastic pressure changes. Therefore, the warning level for this task needs to be reassessed. Feedback indicates that the temperature is still rising rapidly and the heavy rain has not yet subsided. The warning result will be iteratively adjusted based on current data. For example, if the initial warning was "moderate risk," but due to changes in environmental factors, the warning level for Area A is expected to be raised to "risk." Next, the risk assessment result is recalculated based on changes in real-time data. This process includes real-time monitoring of external environmental data such as temperature, humidity, and wind speed, and updating the risk assessment value for Area A based on this data. This includes data on temperature changes over the past two hours. If the temperature rises from 35°C to 40°C, while the environment in region B remains stable with a temperature change of only ±2°C, the risk assessment level of region A will be adjusted based on these real-time changes. Region A will be considered to have a higher risk level, while the risk in region B will remain lower. Then, the feedback information will be applied to the priority assessment of remediation tasks. If the urgency of the remediation tasks in region A increases due to environmental changes, the priority of the remediation tasks will be readjusted based on the current warning results. For example, if the original priority of the remediation task in region A was 3 (routine remediation), the priority of the remediation task will be adjusted to 1 (urgent remediation) due to the impact of temperature fluctuations and pressure changes. Similarly, if the task in region B is not significantly affected by environmental changes, its remediation task priority is estimated to remain at 3, indicating routine remediation, and the final risk warning result will be obtained.

[0039] The above description is merely a specific embodiment of the present invention, but the scope of protection of the present invention is not limited thereto. Any variations or substitutions that can be easily conceived by those skilled in the art within the technical scope disclosed in the present invention should be included within the scope of protection of the present invention. Therefore, the scope of protection of the present invention should be determined by the scope of the claims.

Claims

1. A highway guardrail health monitoring and early warning system based on big data, characterized in that, The system comprises: The sensing data acquisition and monitoring module acquires temperature, pressure, displacement, vibration data, sets the sampling frequency, rejects abnormal data, suppresses interference through low-pass filtering, and obtains health monitoring data; The space-time data fusion and health state determination module determines the guardrail health state based on the health monitoring data, analyzes the regional characteristics through spatial distribution and weight information, and obtains the guardrail health determination state; The damage development trend monitoring module analyzes the health changes of the monitoring area, identifies the trend difference, circumscribes the damage aggravation area, evaluates the influence of external factors on the damage, determines the damage risk, and obtains the damage development trend result based on the guardrail health determination state; The repair priority ranking module evaluates the environmental impact from the perspective of traffic safety, analyzes the emergency repair task according to the damage degree, environmental impact weight, and repair resources, and obtains the repair priority result based on the damage development trend result; The risk warning and response module judges whether to start emergency response and issue a warning based on the repair priority evaluation task urgency, risk level, resources, and environment, re-evaluates the risk, and obtains the risk warning result.

2. The big data based highway guardrail health monitoring and early warning system according to claim 1, characterized in that: The health monitoring data includes temperature data, pressure data, displacement data, and vibration data, the data completeness and continuity includes data missing segments and abnormal segments, the noise reduction effect of the filtered data includes noise elimination and interference weakening, the guardrail health determination state includes health state and monitoring area health state, the health determination analysis includes the influence of regional data on monitoring results, data weight distribution, data mapping, and time and space information alignment, the damage development trend result includes damage aggravation area, health change trajectory obtained through time series analysis, damage risk calculated based on a multivariate regression model, and specific influence mechanism of external factors on damage obtained through environmental parameter analysis, the repair priority result includes repair task order, repair task urgency, repair demand analysis, and repair task priority, and the risk warning result includes warning notification, emergency response start, risk evaluation, and feedback information re-evaluation.

3. The big data based highway guard rail health monitoring and early warning system according to claim 1, characterized in that: The mapping area refers to the space-time data fusion and health state determination module, which identifies the monitoring data belonging to the area according to the geographical characteristics, and corresponds the monitoring data to the specific geographical range according to the positional relationship; The data weight refers to the health state determination process, which is a corresponding proportion for different monitoring data according to the sampling frequency, signal integrity, and regional importance.

4. The big data based highway guard rail health monitoring and early warning system according to claim 1, characterized in that: The trend difference refers to the damage development trend monitoring module, which identifies the direction and amplitude difference of health changes by comparing the health data results of different time periods, and thus determines the development trend of guardrail damage; The emergency response refers to the risk warning and response module, which starts the emergency response process according to the set standard, issues a warning notification, and organizes the implementation of on-site response and emergency disposal measures when the risk evaluation result shows that there is an emergency hidden danger.

5. The big data based highway guard rail health monitoring and early warning system according to claim 1, characterized in that: The sensing data acquisition and monitoring module comprises: The data acquisition sub-module acquires temperature data, pressure data, displacement data and vibration data of the guardrail, sets a sampling frequency according to a difference in environmental parameter changes, performs data acquisition, synchronizes data according to time and space sequences, and obtains original sensing data; The data cleaning sub-module judges the integrity and continuity of data based on the original sensing data, removes missing data and abnormal segments, reduces noise and interference through filtering, and obtains cleaned data streams; The filtering processing sub-module analyzes health state information of the guardrail based on the cleaned data streams, monitors dynamic changes and identifies key features, analyzes temperature, pressure, displacement, vibration and other parameters, and obtains health monitoring data.

6. The big data based highway guard rail health monitoring and early warning system of claim 1, wherein: The space-time data fusion and health state judgment module comprises: The time synchronization sub-module synchronizes each item of data according to a time stamp based on the health monitoring data, calibrates time information of different data sources, compares and analyzes data items within a unified time range, and obtains time-aligned data; The regional data mapping sub-module identifies a monitoring region corresponding to each item of data based on the time-aligned data, maps data to the corresponding region through regional features, and points data to the corresponding region according to geographical positions, and obtains regionalized monitoring data; The health state judgment sub-module assigns a weight to each item of data based on the regionalized monitoring data, analyzes the influence of regional weights on health monitoring results, judges the health state of each monitoring region through comparison of multi-region data, and obtains a health judgment state of the guardrail.

7. The big data based highway guard rail health monitoring and early warning system according to claim 6, characterized in that: The time synchronization sub-module calibrates a time stamp based on health monitoring data, specifically by comparing multiple items of data through a same time reference, identifying time differences between different data sources according to comparison results, and correcting time sequences of data according to the time differences; The regional data mapping sub-module identifies a region based on time-aligned data, specifically by updating data according to a spatial adjacency relationship when position information of monitoring data exceeds a boundary of regional features; The health state judgment sub-module analyzes regionalized monitoring data, specifically by setting data weights according to sampling frequencies and signal integrity, and assigning roles of different regional data in overall evaluation through a weight proportion relationship.

8. The big data based highway guard rail health monitoring and early warning system according to claim 1, characterized in that: The damage development trend monitoring module comprises: A health change trend identification sub-module extracts health data of different time periods based on the health judgment state of the guardrail, analyzes health changes of monitoring regions, compares judgment results of different time periods, identifies change trends according to data differences, and obtains health change trend information; A damage aggravation region delineation sub-module identifies and delineates a region where damage is aggravated based on the health change trend information, analyzes health changes of multiple regions, and obtains damage aggravation region data; A damage risk assessment sub-module analyzes influences of external factors on damage changes based on the damage aggravation region data, evaluates a health change trajectory, judges damage risks according to analysis results, and obtains damage development trend results.

9. The big data based highway guard rail health monitoring and early warning system of claim 1, wherein: The repair priority sorting module comprises: The damage emergency evaluation submodule evaluates the damage level and development rate of the region based on the damage development trend result, determines the urgency of the repair task by comparing the damage aggravation degree, and obtains environment data; The environmental impact analysis submodule analyzes the influence of external factors on damage change based on the environment data, judges the influence of environmental factors on repair demand through damage development trend data, and obtains environmental factor influence result; The repair task sequencing submodule sets the task sequence based on the environmental factor influence result, analyzes the emergency degree, influence factor and repair priority of the task, sets the priority of the repair task, and obtains repair priority result.

10. The big data based highway guard rail health monitoring and early warning system according to claim 1, characterized in that: The risk warning and response module comprises: The risk evaluation submodule analyzes the correlation data of each repair task based on the repair priority result, evaluates the risk of the task, analyzes the emergency and available response time of the task, determines the risk level of the task execution, and obtains risk evaluation result; The emergency response determination submodule determines whether to start the emergency response process based on the risk evaluation result and the emergency degree of the task, analyzes the repair task that needs to be processed immediately according to the set response standard, and obtains the determination result of whether to start the emergency response; The warning feedback processing submodule reevaluates the warning based on the determination result of whether to start the emergency response and feedback information, changes the risk evaluation result through real-time data, and adjusts the repair task priority according to the feedback information, and obtains risk warning result.

Images